Date of Award

Fall 2025

Rights

Access is available to all users

Document Type

Thesis

Degree Name

Master of Science (MS) in Biology

Department

Biology

Abstract

Characterizing mosquito species composition and seasonal dynamics is fundamental to understanding how mosquito communities change across time and space. This study examined the structure, temporal patterns, and spatial variation of the mosquito community at Turnbull National Wildlife Refuge (TNWR) using three years of CDC light-trap data (2017–2019). We applied multiple complementary analytical approaches—including presence/absence and CPUE models, annual and species-specific generalized additive models (GAMs), peak-timing ANOVA, and multivariate analyses (NMDS, PERMANOVA, and indicator species analysis)—to quantify species-level seasonality, phenological staggering, and community-level patterns across years and between two sampling locations (Gate and Headquarters). Across all model families, species identity and time were the dominant drivers of variation, while spatial differences were present but comparatively modest and restricted to a subset of taxa. Annual and species-specific GAMs revealed that each species exhibited distinct seasonal curves with substantial interannual variability, and peak-timing analysis confirmed clear staggering between early-season Aedes species and later-season Anopheles, Coquillettidia, and Culex. Multivariate analyses further showed that year and location explained most community variation, with location contributing smaller but significant effects. Although not a primary objective, these findings also provide ecological context relevant to vector surveillance by identifying dominant species, key seasonal windows, and temporal peaks in the community. Together, the results demonstrate that mosquito seasonality at TNWR reflects strongly species-specific and temporally dynamic patterns, emphasizing the importance of detailed community-level analyses for understanding mosquito ecology in temperate regions.

Share

COinS